www.elsevier.com/locate/brainres

Brain Research 1044

Research report

Electroacupuncture attenuates morphine withdrawal signs and c-Fos

expression in the central nucleus of the amygdala in freely moving rats

Sheng Liua, Wenhua Zhoub,T, Huifen Liub, Guodong Yangb,T, Weikang Zhaoa

aShanghai University of Traditional Chinese Medicine, Shanghai, 200032, P.R. ChinabNingbo Addiction Research and Treatment Center, 42, Xibeijie Street, Ningbo 315010, P.R. China

Accepted 23 February 2005

Available online 15 April 2005

Abstract

Experimental efforts for understanding the mechanisms of electroacupuncture (EA) for opiate addiction are partially hampered by restraint

stress. In unrestrained animals, it is difficult to perform EA stimulation at acupuncture points frequently selected on the four limbs. The

present study was performed to evaluate the effect of EA at the acupuncture point Shen-Shu (BL.23) on morphine withdrawal signs and c-Fos

expression of the amygdala in freely moving rats or restrained rats. We applied immunohistochemistry to detect c-Fos-positive nuclei.

Corticosterone levels and behavioral responses were measured during EA stimulation. The needles were bilaterally inserted and fixed at

BL.23, and 100-Hz electric stimulation was conducted 30 min before naloxone-precipitated withdrawal. In both freely moving rats and

restrained rats, EA significantly reduced the signs of morphine withdrawal. Notably, EA stimulation in freely moving rats attenuated c-Fos

expression in the central nucleus of the amygdala while EA in restrained animals increased this response. In addition, the restrained rats

emitted greater levels of vocalization and facial expression than freely moving rats during EA stimulation. Corticosterone levels were also

significantly higher in restrained animals after EA stimulation. The new EA paradigm demonstrated in the present study might help the

analysis of certain physiological responses induced by EA that would otherwise have been hindered by restraint stress.

D 2005 Elsevier B.V. All rights reserved.

Theme: Neural basis of behavior

Topic: Drugs of abuse: opioids and others

Keywords: Acupuncture point BL.23; Morphine withdrawal; The amygdala; c-Fos immunohistochemistry; Restraint stress

1. Introduction

Abrupt cessation of chronic opiate use results in a

characterized withdrawal syndrome that includes nausea,

dysphoria, and anxiety [23]. These consequences of

abstinence are thought to be important factors contributing

to opiate addiction. Acupuncture and electroacupuncture

(EA) have been applied with great success to attenuate

behavioral signs of morphine withdrawal in addicts

[8,30,39]. The effects of acupuncture on drug addiction

have also been verified by animal experiments. The with-

0006-8993/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.brainres.2005.02.075

T Corresponding authors. Fax: +86 574 87345976.

E-mail address: liusheng@nartc.cn (G. Yang).

drawal syndrome observed in morphine-dependent rats can

be effectively suppressed by 100-Hz EA [20,47]. Morphine-

induced conditioned place preference can be successfully

suppressed by 2- or 100-Hz EA [38,42]. These animal

studies have provided important information for under-

standing the underlying neurobiological mechanisms of

acupuncture and EA in the treatment of opiate addiction.

However, experimental efforts are, at least partially, ham-

pered by several limitations of laboratory animals. The

most important one is that laboratory animals must be

restrained for several minutes to complete acupuncture

treatment. This has the following inherent problems: (1)

acute restraint stress itself can interfere with acupuncture

stimulation [27]; (2) both a single restraint session and a

repeated restraint stress similarly enhance the effects of

(2005) 155–163

S. Liu et al. / Brain Research 1044 (2005) 155–163156

morphine on locomotor activity [37]. Those rats with high

locomotor activity are more vulnerable to drug addiction

[34]; (3) restraint stress induces the release of endogenous

opioids [3] and produces opioid-like effects [2,32]; (4)

restraint stress can activate several brain regions [10],

representing a major confounding variable for the assess-

ment of EA-induced c-Fos expression in brain areas.

In animal studies, the acupuncture points most frequently

selected are Zusanli (ST.36) and Sanyinjiao (SP.6) located

on the leg. It is difficult to perform the needle manipulation

at these acupuncture points, which are located on the four

limbs, in unrestrained animals. In fact, according to the

theory of traditional Chinese medicine, some acupuncture

points represent discrete locations on the body, where

manual or electrical stimulation can exhibit similar, if not

identical, therapeutic effects [4]. This raises the possibility

that some acupuncture points located on the back and head

have similar therapeutic effects as Zusanli (ST.36) and

Sanyinjiao (SP.6). An important acupuncture point, Shen-

Shu (BL.23), is located on the back and is commonly used

for analgesia and sedation in our clinic. It has been shown

previously that EA at BL.23 has similar antinociceptive

effects as EA at ST.36 [4,17]. We hypothesized that some

acupuncture points that induce analgesia could play a role in

the treatment of opioid addiction.

The goals of our present study were to evaluate the effect

of EA at BL.23 on morphine withdrawal syndrome in both

freely moving and restrained rats. We compared the

behavioral responses during EA stimulation. To determine

whether EA exhibits different activation within certain brain

regions in restrained or unrestrained rats, we initially applied

c-Fos immunomapping to investigate functional activation

of the amygdala during precipitated withdrawal from

morphine and EA stimulation. The amygdala is thought to

be important in mediating the behavioral, autonomic, and

endocrine responses to stressors [11,12,22]. Functional

neuroimaging studies demonstrate that the human amygdala

is activated during negative affective states [11]. Further-

more, the amygdala is well known to be involved in opioid

withdrawal [23,28]. A number of studies have shown that

both morphine withdrawal and restraint stress induce the

activation of c-Fos in the amygdala [5,16,18,35,41]. By

excluding or minimizing interference from restraint stress,

the present study assessed the brealQ effects of EA stimu-

lation on c-Fos expression of the amygdala in morphine

withdrawal rats.

2. Materials and methods

2.1. Subjects

Male Sprague–Dawley rats (250–300 g) from the Zhe-

jiang Center of Experimental Animals were used. They were

randomly assigned and housed collectively (four per cage)

under controlled environmental conditions (22 8C, 12-h

light/dark cycle) with free access to food and water. All

animal treatments were performed in strict accordance with

the National Institutes of Health Guide for the Care and Use

of Laboratory Animals. All experiments were conducted

during the light cycle.

2.2. Drug and withdrawal behavioral scores

Morphine treatment and precipitation of withdrawal were

performed according to published procedures [48]. Rats

were subcutaneously injected with morphine–HCI twice

daily. The dose of each injection was 10 mg/kg in the first

day and increased by 10 mg/kg each day thereafter (10, 20,

30, 40, 50 mg/kg). On day 6, morphine withdrawal

syndrome was precipitated by an intraperitoneal (ip)

injection of 4 mg/kg naloxone hydrochloride (Sigma,

USA), 4 h after one single injection of 50 mg/kg morphine.

The withdrawal behavioral scores were calculated during

the following 45 min (divided into three 15-min periods for

scoring) in a quiet room by one observer who did not know

what experimental treatment had been administered. The

scored signs are based on published criteria [26,36,48], with

minor modifications. The scores obtained for each 15-min

period were noted on the score sheet and were summed over

the 45-min observation period. Withdrawal scores included

teeth chattering, wet dog shake, diarrhea, irritability,

salivation, and abnormal posture including writhing, dig-

ging, and hunching. The amount of weight loss experienced

by the animal during the observation period was also

analyzed. Weight loss during the 45-min observation period

was scored as 1 for a loss of b2%, 5 for a loss of b4%, 10

for a loss of b6%, 15 for a loss of b8%, and 20 for a loss

of N8%.

2.3. EA stimulation

Electric stimulation was conducted for 30 min before

naloxone-precipitated withdrawal. Stainless steel needles

were bilaterally inserted to a depth of 5 mm into BL.23 (one

to two rib’s width lateral to the caudal border of the spinous

process of the second lumbar vertebra) or ST.36 (located

near the knee joint, between the muscle anterior tibialis and

muscle extensor digitorum longus). Constant current square-

wave electric stimulation produced by an electroacupunc-

ture apparatus (Model G-6805-2, Shanghai Medical Elec-

tronic Apparatus, China) was administered via the two

needles. The frequency of stimulation used was 100 Hz

(0.2 ms pulse width). The intensity of the stimulation was

increased stepwise from 1.5 to 2 mA, with each step lasting

for 15 min.

2.4. c-Fos immunohistochemistry

Rats were deeply anesthetized with sodium pentobarbital

(60 mg/kg, ip) and killed by transcardial perfusion of

200 ml saline followed by 200 ml 4% paraformaldehyde in

Fig. 1. Photograph of the electroacupuncture procedures and placement of

acupuncture needles. Two needles were fixed bilaterally at acupuncture

point BL.23. The needles were attached to an electrical stimulator for EA

treatment. The rat could move freely in an observation chamber during EA

stimulation.

S. Liu et al. / Brain Research 1044 (2005) 155–163 157

0.1 mol/L phosphate buffer (PB). Brains were dissected and

postfixed in the same fixative and then stored in 30%

sucrose at 4 8C for 3–5 days. Coronal sections (30 Am in

thickness; �2.56 mm from bregma according to the atlas of

Paxinos and Watson [33]) were cut on the cryostat at

�25 8C. Sections were rinsed in 0.01 M phosphate-buffered

saline (PBS) and incubated in PBS containing 5% normal

goat serum and 0.3% Triton X-100 for 30 min, and then in

Fos antibody (rabbit polyclonal antibody. Santa Cruz, USA)

diluted at 1:200 in PBS at 4 8C for 48 h. After rinsing three

times with PBS, sections were incubated in the biotinylated

goat anti-rabbit secondary antibody (Sigma, USA, diluted

1:200 with PBS) for 2 h and washed again. Then all

corresponding sections were placed in the avidin-biotin-

peroxidase complex solution for 60 min. Finally, DAB

was used for visualization of Fos immunoreactivity. The

reaction was stopped by several PBS washes. Sections were

then mounted on gelatin-coated slides, air-dried, dehydrated

through graded alcohols, cleared in xylene, and cover-

slipped with Eukitt.

2.5. Quantification of Fos-positive nuclei in the amygdala

The amygdala is divided into a number of subregions.

Our study focused on two subregions: (1) the basolateral

amygdala (BLA)—this region is comprised of principal

output neurons that are glutamatergic, pyramidal-shaped

neurons and provide excitatory output to a number of rostral

brain regions; (2) the central nucleus (CeA)—this region

contains primarily GABAergic neurons that project to

brainstem structures, including the locus caeruleus and

hypothalamus. The two subregions have been proposed to

play an important role in the expression of many withdrawal

signs [23,28]. Sections were scanned using an Olympus

BX51 microscope. Image analysis was carried out with the

aid of an image analysis system (Microimage, Olympus

Optical (Europa), Hamburg, Germany. Three consecutive

sections were taken from each animal, and the Fos-positive

nuclei were counted bilaterally, based on a randomization

procedure. A computer-generated rectangle (250 � 600 Am)

was placed in a fixed area of the BLA and CeA of each

section, and the analysis software counted stained nuclei

within the area (see Fig. 5F for specific area used for CeA

and BLA).

2.6. Experimental protocols

In experiment 1, to determine whether BL.23 has similar

therapeutic effects as ST.36 on naloxone-precipitated with-

drawal signs, rats were restrained and submitted to EA

bilaterally at BL.23 or ST36 30 min before naloxone-

precipitated withdrawal on day 6. Rats merely restrained

and not administered EA, and rats that received no treatment

before morphine withdrawal were used as controls.

In experiment 2, to observe the effect of EA at BL.23 on

morphine withdrawal and c-Fos expression in the amygdala

in restrained or freely moving rats, rats were divided into

four groups (n = 7): (1) freely moving and EA group: on day

5, rats were anesthetized with light halothane. Two stainless-

steel needles were inserted at the acupuncture point BL.23

on both sides of the spine and fixed in place with

cyanoacrylate glue and acrylic dental cement. On day 6,

30 min before naloxone-precipitated withdrawal, the nee-

dles were connected with the output terminals of the

stimulator. Freely moving rats received electric stimulation

in an observation chamber (21 � 28 � 20.5 cm) as shown in

Fig. 1. (2) Sham-EA group: the needles were fixed at BL.23

on day 5, but electric stimulation was not conducted on day

6. (3) Restrained and EA group: rats were restrained and

submitted to EA at BL.23 on day 6. (4) The morphine

withdrawal group: EA was not administered before nalox-

one-precipitated morphine withdrawal. Considering that it

was difficult to perform the needle manipulation at ST.36 in

freely moving rats, we did not run the group restrained and

ST.36 as control in experiment 2.

After calculating withdrawal scores, four rats from each

group were randomly selected for c-Fos immunohistoche-

mistry. Four additional morphine-dependent rats (that

received an isotonic saline injection but not naloxone on

day 6) were used for studying the expression of c-Fos in the

amygdala only.

In experiment 3, we observed different behavioral res-

ponses in restrained rats and in freely moving rats during

EA stimulation. In order to exclude the possible effects of

morphine dependence on the behavioral observation, we

used 12 opiate-naive rats. Six rats were restrained and

submitted to EA at BL.23, while six freely moving rats

received EA as described in experiment 2. Vocalization and

facial expression were measured during EA stimulation.

Both vocalization and facial expression in which the corners

of the mouth are retracted, resulting in exposure of the teeth,

were recorded on videotape. A standard scoring system was

used to analyze the presence or absence of vocalization or

the facial expression during EA stimulation. After EA

stimulation, animals were decapitated rapidly. Trunk blood

samples (500 Al in heparinized tubes) were collected and

S. Liu et al. / Brain Research 1044 (2005) 155–163158

centrifuged for 5 min at 4000 rpm. The plasma was

immediately frozen on dry ice and kept frozen at �20 8Cuntil measurement of corticosterone. A radioimmunoassay

for plasma corticosterone was conducted with the help of

Dr. Sun using a kit from ICN Biochemicals (Costa Mesa,

CA) and 125I-corticosterone as the tracer. An additional

group of six opiate-naive rats without any treatment was

used as a control for corticosterone response.

2.7. Statistical analysis

Data from experiments 1 and 2 were analyzed by one-

way or two-way analysis of variance (ANOVA). When

significance was found using ANOVA procedures, Fisher’s

LSD post hoc testing was used to compare individual

treatment groups. Values are expressed as mean F SEM. In

experiment 3, similar statistics were performed on the

corticosterone levels. Fisher’s exact test was used to analyze

the presence and absence of vocalization and facial ex-

pression during EA stimulation. P b 0.05 was considered

statistically significant.

3. Results

3.1. Effects of EA at BL.23 acupuncture point on

naloxone-precipitated withdrawal syndrome

In experiment 1, EA at both BL.23 and ST.36 conside-

rably attenuated opiate withdrawal behaviors. As depicted

in Fig. 2, significant attenuation was seen for the total score

(F(3,24) = 45.36; P b 0.001), weight loss (F(3,24) = 14.53;

Fig. 2. Effect of EA stimulation on naloxone-precipitated withdrawal in

morphine-dependent rats. The restraint rats received EA stimulation at

ST.36 (n = 7) or at BL.23 (n = 7). The restraint rats without EA (n = 7)

were as control. No difference was observed in the severity of opiate

withdrawal behaviors in restraint animals compared with morphine with-

drawal rats (n = 7). EA at both BL.23 and ST.36 considerably attenuated

opiate withdrawal behaviors. Data are expressed as mean values F SEM.

Statistically different from morphine withdrawal rats (*P b 0.05), analysis

of variance followed by Fisher’s LSD test for multiple comparisons.

P b 0.001), salivation (F(3,24) = 18.15; P b 0.001),

diarrhea (F(3,24) = 8.01; P b 0.05), wet dog shake

(F(3,24) = 3.97; P b 0.05), and abnormal position

(F(3,24) = 8.61; P b 0.01). There was a non-significant

trend for EA at BL.23 and ST.36 for a decrease in the scores

of teeth chattering (F(3,24) = 0.094; P = 0.96). EA at BL.23

and ST.36 exhibited a similar suppression of opiate with-

drawal syndrome. There were no significant differences

between EA at BL.23 and EA at ST.36 in their attenuation

of the total score, weight loss, salivation, diarrhea, wet dog

shake, and other abnormal position, as determined by

Fisher’s LSD post hoc test. In addition, no difference was

observed in the severity of opiate withdrawal behavior in

animals that were merely restrained but not administered EA

before morphine withdrawal, as compared with the mor-

phine withdrawal group.

3.2. Effect of EA at BL.23 on naloxone-precipitated

withdrawal in freely moving rats

As shown in Fig. 3, EA in freely moving rats or in

restraint rats produced significant reductions in several

withdrawal behaviors, including total signs, weight loss,

salivation, diarrhea, and abnormal behaviors, as compared

with the morphine withdrawal group and Sham-EA group.

Although there was a trend for EA to decrease the scores of

wet dog shakes in experiment 2, this effect did not achieve

statistical significance (F(3,24) = 1.54; P N 0.05). There

was a non-significant trend for EA to decrease the scores of

teeth chattering (F(3,24) = 0.81; P = 0.48). Of note, no

significant differences were found between rats without

electric stimulation (where the needles were fixed at the

acupuncture points only) and morphine withdrawal rats.

Overall, differences were observed between the four groups

for total score (F(3,24) = 11.05; P b 0.001), weight loss

(F(3,24) = 3.67; P b 0.05), salivation (F(3,24) = 17.01; P b

0.001), diarrhea (F(3,24) = 3.97; P b 0.05), and abnormal

position (F(3,24) = 12.42; P b 0.001).

3.3. Effect of EA at BL.23 on c-Fos expression in the

amygdala in morphine withdrawal rats

Coronal sections of the brain were obtained and Fos-

positive nuclei were imaged in the BLA and CeA of the

different groups (Figs. 4 and 5). As illustrated in Fig. 4, in

morphine-dependent rats, only a few c-Fos positive neurons

could be detected in the CeA and BLA (Fig. 5E). By

contrast, morphine withdrawal increased Fos immunoreac-

tivity in the two components of the amygdala (Fig. 5A). The

CeA displayed a strong increase in the number of c-Fos

positive neurons whereas BLA displayed fewer activated

neurons, even if the changes were clearly significant

compared with morphine-dependent rats (Fig. 4). In the

BLA, there were no differences in Fos expression among the

freely moving and EA group, Sham-EA group, restrained

and EA group, and the morphine withdrawal group, as

Fig. 4. Quantitative analysis of Fos-positive nuclei in the BLA and CeA. In

the CeA, the restrained and EA group exhibited a significant increase in

Fos-positive nuclei (*P b 0.05 versus the morphine withdrawal group). EA

stimulation in freely moving rats attenuated c-Fos expression in the CeA

(#P b 0.05 versus the morphine withdrawal group and Sham-EA group). In

the BLA, no significant differences were found, except in the morphine-

dependent rats. Data are expressed as mean values F SEM.

Fig. 3. Effect of EA at BL.23 on morphine withdrawal in freely moving

rats. Freely moving rats (n = 7) or restraint rats (n = 7) received EA

treatment before naloxone-precipitated withdrawal. Rats without electric

stimulation (only were the needles fixed at the acupuncture points) (Sham-

EA group, n = 7) and morphine withdrawal rats (n = 7) were set as controls.

Data are expressed as mean values F SEM. Statistically different from

morphine withdrawal rats (*P b 0.05), analysis of variance followed by

Fisher’s LSD test for multiple comparisons.

S. Liu et al. / Brain Research 1044 (2005) 155–163 159

determined by Fisher’s LSD post hoc test (P N 0.05).

ANOVA demonstrated significant differences among treat-

ment groups in the number of Fos-positive nuclei in the

CeA (P b 0.001). Of note, post hoc comparisons indicated

that the restrained and EA group exhibited an increase in

Fos-positive nuclei in the CeA, whereas EA stimulation in

freely moving rats attenuated c-Fos expression relative to

the morphine withdrawal group (Fisher’s LSD test; P b

0.05; Figs. 5B and D). No differences in Fos expression

were noted between the Sham-EA group and morphine

withdrawal group (P N 0.05; Fig. 5C).

3.4. Vocalization and facial expression during EA

stimulation

Behavioral indicators of acute stress are characterized

by vocalization and facial expression. As summarized in

Table 1, the freely moving rats emitted significantly fewer

vocalizations than the restrained rats during needle

manipulation (P b 0.05). In addition, the freely moving

animals engaged in less facial expression compared with

the restrained animals during EA stimulation (Fisher’s

exact test; P b 0.05).

3.5. Corticosterone secretion after EA stimulation

Corticosterone secretion differed among the three groups

after EA stimulation (ANOVA; F(2,15) = 39.17; P b 0.001;

Fig. 6). Corticosterone levels were significantly higher in

the restraint animals than in the freely moving animals

(Fisher’s LSD test; P b 0.01). Compared with the control

group, the freely moving rats receiving EA stimulation also

exhibited a higher corticosterone response (Fisher’s LSD

test; P b 0.001).

4. Discussion

The major results of this study can be summarized as

follows. EA at BL.23 in freely moving rats or in restrained

rats produced significant reductions in several withdrawal

behaviors, including total signs, weight loss, salivation,

diarrhea, and abnormal behaviors. Notably, EA stimulation

in freely moving rats attenuated c-Fos expression in the

CeA, which would otherwise have been hindered by

restraint stress. The restrained rats emitted much more

vocalization and facial expression than the freely moving

rats during EA stimulation. Corticosterone levels were

higher in the restraint animals after EA stimulation.

Accordingly, the new EA paradigm described in the present

study might help the analysis of certain physiological

responses induced by EA.

According to our clinical experience and the theory of

traditional Chinese medicine, 100-Hz EA at BL.23 can be

used to treat acute or chronic pain [4]. Han and his

colleagues have made a detailed survey on the analgesic

effect of 100-Hz EA. They found that analgesia induced by

100-Hz EA resulted from accelerating the release of

dynorphin from the spinal cord of the rats [6,15,19]. In

accord with this was the finding that the analgesic effect of

100-Hz EA observed in morphine-dependent rats could be

blocked by a high dose of naloxone only [47]. On the other

hand, dynorphin has been shown to be the endogenous

ligand of the n-opioid receptor. Activation of n-opioidreceptors in the spinal cord suppresses morphine withdrawal

Fig. 5. Representative coronal sections showing Fos immunoreactivity in the amygdala. Morphine withdrawal rats (A); EA stimulation in the restraint rats (B);

Needles were fixed without electric stimulation in freely moving rats (C); EA stimulation in freely moving rats (D); and morphine-dependent rats (E).

(F) Schematic coronal section through the amygdala illustrated placement of the 200 � 500 Am rectangle in the BLA and CeAwhere Fos-positive nuclei were

counted. Number at the top left of F represents the distance (in millimeters) from bregma. The depicted coronal section was adapted from the atlas by Paxinos

and Watson [33]. Scale bar, 200 Am.

S. Liu et al. / Brain Research 1044 (2005) 155–163160

syndrome and blockade of spinal n-opioid receptors

aggravates the syndrome [9]. Wen et al. reported suppres-

sion of heroin withdrawal syndrome by dynorphin A (1–17)

in 1982 [44] and dynorphin A (1–13) in 1992 [45]. As

would thus be expected, 100-Hz EA at some acupuncture

points used for analgesia could be applied to the treatment

Table 1

Behavioral responses to EA stimulation

Freely moving

EA (n = 6)

Restrained

EA (n = 6)

P

Vocalization 0 4 b0.05

Facial expression 1 5 b0.05

Behavioral responses in freely moving rats and restraint rats during EA

stimulation are displayed. Significantly fewer freely moving rats engaged in

vocalization and facial expression than did the restraint animals as

determined by Fisher’s exact test.

of opioid addiction. Our present study showed for the first

time that EA applied at the BL.23 acupuncture point

could effectively suppress withdrawal syndrome in either

restrained or freely moving rats, although slight differences

in wet dog shakes were observed in experiment 1 and

experiment 2. Clinically, BL.23, tested in the present

study, could provide us with a new selection of effective

acupuncture points for successful treatment of morphine

withdrawal syndrome. Further studies on the synergistic

combination of BL.23 and other acupuncture points, such as

Zusanli (ST.36), could assist acupuncturists to use a

balanced and appropriate choice for combining points in

the treatment of addicts.

Opiate withdrawal is a particularly complex syndrome.

The amygdala may induce or modulate opiate withdrawal

symptoms [18,21,43]. One means of assessing functional

activity in the brain is to study the expression of the

Fig. 6. Plasma corticosterone secretion in response to a 30-min EA

stimulation. Corticosterone levels were significantly higher in the restraint

rats than in the freely moving rats after EA stimulation. Compared with the

control group, the freely moving rats receiving EA stimulation also

exhibited higher corticosterone response as determined by Fisher’s LSD

test. **P b 0.01; ***P b 0.001.

S. Liu et al. / Brain Research 1044 (2005) 155–163 161

immediate early gene c-fos and its protein product c-Fos,

which is expressed as a function of cellular activation

following extracellular stimuli [24,31]. A number of studies

have shown that several components of the amygdala

complex are selectively activated during drug withdrawal

[5,16,18,41]. The increase of c-Fos-positive neurons in the

CeA and BLA after morphine withdrawal reported here is

consistent with previous studies [16,41]. It is notable in our

present study that EA in freely moving rats attenuated

elevated c-fos expression in the CeA during morphine

withdrawal. Our current findings favor the hypothesis that

the anti-withdrawal effects of EA at BL.23 were mediated

partially by suppressing neuronal hyperexcitability in

the amygdala. Of course, further studies must be performed

to clarify this issue. On the other hand, some studies

emphasize that the motivational components of opiate

withdrawal appear to be centrally mediated by limbic

structures such as the nucleus accumbens and amygdala

[16,21,28,40]. The CeA and BLA have been extensively

and differentially involved in associative learning processes,

attributing affective salience to environmental stimuli paired

with drug effects [14]. Therefore, elevated c-fos expression

in the CeA might be associated with the motivational

components of opiate withdrawal. Our observation that EA

stimulation in freely moving animals suppressed elevated

c-fos expression in the CeA indicated that EA at BL.23

might have some therapeutic benefit in the treatment of the

negative motivational aspect of opiate withdrawal. Interest-

ingly, in contrast to EA stimulation in freely moving rats,

the restrained and EA group exhibited an increase in c-fos

expression in the CeA; considering our observation that

restraint stress increased the c-Fos expression of CeA in

morphine withdrawal rats (data not shown). The results

from the restrained and EA group might be because a

restraint stress itself can activate neurons and induce c-fos

expression in the CeA. The amygdala is known to be a

critical area involved in mediating the behavioral, auto-

nomic, and endocrine responses to stressors [11,12,22]. The

motivational or emotional state of the animal can interfere

with c-Fos expression in the amygdala [25]. Therefore, in

restrained rats, the effects of EA on c-Fos expression in the

amygdala might be confounded by restraint stress itself or

the motivational or emotional state induced by stress.

Laboratory practices such as animal handling, injections,

and the introduction of animals to novel environments are

known benign stressors that are likely to influence behavior

and the release of steroids and endogenous opioids [1].

Previous studies have shown that aversive stressors (e.g.,

swim stress and foot-shock) can influence withdrawal

phenomena. Christie and Chesher [7] reported that in mice

chronically exposed to swim stress, naloxone induces

withdrawal behaviors that closely resemble those produced

following chronic morphine treatment, while Williams et al.

[46] have shown that the stress of prior exposure to foot-

shock augments withdrawal responses precipitated by

naloxone in morphine-dependent rats. In contrast to both

of the above experiments, our present experiment indicated

that the behavioral signs of morphine withdrawal induced in

restrained and unrestrained animals were statistically indis-

tinguishable. Both restrained rats and freely moving rats

pretreated with EA before naloxone administration showed

few signs of an abstinence syndrome. The discrepancy

between these studies and the present one might be

explained by differences in the stressors. Swim stress and

foot-shock are much more severe than a single restraint

stressor. It is well documented that withdrawal behavior was

generally greater in the opiate-dependent animals in the

bhighQ rather than the blowQ stress group [35]. Nevertheless,

the possibility that the bstressQ of normal laboratory restraint

procedures might influence behavior cannot be discounted.

In our present study, the restrained rats emitted more

vocalization and facial expression than the freely moving

rats. Corticosterone levels were higher in the restraint

animals. Further, the level of Fos protein expression in the

amygdala was significantly increased in restrained rats

compared with that of the freely moving rats. These results

emphasize the importance of minimizing stressful proce-

dures and performing appropriate controls in assessing the

effect of EA on opiate dependence and withdrawal. Some

studies have used new experimental protocols to diminish

the influence of acute restraint stress during needle

manipulation. For example, de Medeiros et al. [13] used a

protocol of repeated immobilization in an attempt to

habituate the animals to this stimulus during EA stimulation.

Shi et al. [38] reported that they chose to use 100-Hz EA or

peripheral electric stimulation at a lower intensity of 0.5–

1.0–1.5 mA in order to eliminate possible stress. Our new

EA paradigm, in which the needles are fixed at the

acupuncture points in the freely moving animals, resulted

in significantly fewer vocalizations and less facial expres-

sion. Further, corticosterone levels were significantly lower

in freely moving rats than in restrained rats after EA

stimulation, indicating that less stress was induced.

S. Liu et al. / Brain Research 1044 (2005) 155–163162

With regard to the EA paradigm in the present study, there

are some variable factors that need to be taken into account.

Firstly, in comparison to the classical acupuncture or EA

paradigms in which EA stimulation only last tens of minutes,

our new EA stimulation paradigm in freely moving rats

seemed to show a long-term effect of acupuncture stimulation

(the needles were inserted and fixed at the acupuncture points

1 day before naloxone-precipitated morphine withdrawal).

We cannot completely exclude the possibility that an

inflammatory reaction was induced in the local insertion of

needle. However, the acupuncture sites had been treated first

with antiseptic solution to avoid contamination of the needle

manipulation, and we did not observe signs of pain in the rats

(such as reluctance to move about, eat or drink, or vocal-

ization), so we used control group Sham-EA. Interestingly,

without electric stimulation, acupuncture only (where the

needles were fixed at the acupuncture points) has no

significant effect on morphine withdrawal. These data

support the clinical and experimental observation that a

therapeutic effect of acupuncture is achieved by sufficient

stimulation intensity. Clinically, acupuncturists think highly

of the typical needling sensation. This sensation (the

characteristic needling sensation at the acupuncture points,

featured by numbness, heaviness, distention, and soreness) is

traditionally believed to be essential in achieving acupunc-

ture’s therapeutic effect. It is elicited by sufficient manual

manipulation (e.g., rotation, up-and-down motion) of the

inserted acupuncture needle. EA stimulation is widely used as

a substitute for classical needle manipulation. In agreement

with previous studies [29,38], results obtained here also

indicate that sufficient electric stimulationmight be necessary

to achieve EA’s therapeutic effectiveness. Thus, our EA

paradigm may still bear significance in mimicking the short-

term acupuncture stimulation. Secondly, it is still controver-

sial as to how to set suitable control stimulation for

acupuncture or EA. The perfect control, of course, is one in

which needles are inserted into non-acupuncture point areas

but with electric stimulation. Unfortunately, such a control is

impractical in rats. It is difficult to select suitable non-

acupuncture point areas on a rat’s back, in which so many

acupuncture points are located. In the present study, we used

Sham-EA (the needles were fixed at the same acupuncture

point in freely moving rats but without electric stimulation). It

is similar to the real EA and could be helpful for explaining

the results (only one stimulation parameter was changed). In

summary, the new EA paradigm described in the present

study might help the analysis of certain physiological res-

ponses induced by EA that would otherwise have been

hindered by restraint stress.

Acknowledgments

We thank F.Q. Zhang for helpful comments on the

manuscript and technical assistance. This work was

supported by National Basic Research program of China

(2003 CB515404) and National Nature Science Foundation

of China (30100051).

References

[1] H. Akil, J. Madden, R.L. Patrick, J.D. Barchas, Stress-induced

increase in endogenous opiate peptides: concurrent analgesia and its

partial reversal by naloxone, in: H.W. Kosterlitz (Ed.), Opiates and

Endogenous Opioid Peptides. Amsterdam; Elsvier North-Holland

Biomedical Press, 1982, pp. 63–70.

[2] S. Amir, Z. Amit, Endogenous opioid ligands may mediate stress-

induced changes in the affective properties of pain related behaviour

in rats, Life Sci. 23 (1978) 1143–1152.

[3] B.D. Appelbaum, S.G. Holtzman, Characterization of stress-induced

potentiation of opioid effects in the rat, J. Pharmacol. Exp. Ther. 231

(1984) 555–565.

[4] Beijing College of Traditional Chinese Medicine, Shanghai College

of Traditional Chinese Medicine, Nanjing College of Traditional

Chinese Medicine, Acupuncture Institute of the Academy of

Traditional Chinese Medicine, Essentials of Chinese acupuncture,

Foreign Language Press, Beijing, People’s Republic of China,

1980.

[5] B. Calvino, J. Lagowska, Y. Ben-Ari, Morphine withdrawal syn-

drome: differential participation of structures located within the

amygdaloid complex and striatum of the rat, Brain Res. 177 (1979)

19–34.

[6] X.H. Chen, J.S. Han, Analgesia induced by electroacupuncture of

different frequencies is mediated by different types of opioid

receptors: another cross-tolerance study, Behav. Brain Res. 47

(1992) 143–149.

[7] M.J. Christie, G.B. Chesher, Physical dependence on physiologically

released endogenous opiates, Life Sci. 30 (1982) 1173–1177.

[8] V. Clement-Jones, L. McLoughlin, P.J. Lowry, G.M. Besser, L.H.

Rees, H.L. Wen, Acupuncture in heroin addicts; changes in Met-

enkephalin and beta-endorphin in blood and cerebrospinal fluid,

Lancet 2 (1979) 380–383.

[9] C.L. Cui, L.Z. Wu, J.S. Han, Spinal k-opioid system plays an

important role in suppressing morphine withdrawal syndrome in the

rat, Neurosci. Lett. 295 (2000) 45–48.

[10] W.E. Cullinan, J.P. Herman, D.F. Battaglia, H. Akil, S.J. Watson,

Pattern and time course of immediate early gene expression in rats

brain following acute stress, Neuroscience 64 (1995) 477–505.

[11] R.J. Davidson, W. Irwin, The functional neuroanatomy of emotion and

affective style, Trends Cogn. Sci. 3 (1999) 11–21.

[12] M. Davis, The role of the amygdala in fear and anxiety, Annu. Rev.

Neurosci. 15 (1992) 353–375.

[13] M.A. de Medeiros, N.S. Canteras, D. Suchecki, L.E. Mello,

Analgesia and c-Fos expression in the periaqueductal gray induced

by electroacupuncture at the Zusanli point in rats, Brain Res. 973

(2003) 196–204.

[14] B.J. Everitt, J.A. Parkinson, M.C. Olmstead, M. Arroyo, P. Robledo,

T.W. Robbins, Associative processes in addiction and reward. The role

of amygdala-ventral striatal subsystems, Ann. N. Y. Acad. Sci. 877

(1999) 412–438.

[15] H. Fei, G.X. Xie, J.S. Han, Low and high frequency electro-

acupuncture stimulations release (Met5) enkephalin and dynorphin

A in rat spinal cord, Chin. Sci. Bull. 32 (1987) 1496–1509.

[16] F. Frenois, M. Cador, S. Caille, L. Stinus, C. Le Moine, Neural

correlates of the motivational and somatic components of naloxone-

precipitate morphine withdrawal, Eur. J. Neurosci. 16 (2002)

1377–1389.

[17] M.X. Gao, The effect of thumb-pressure at Shen-Shu on renal colic,

JCAM 17 (2001) 9–10.

[18] K.N. Gracy, L.A. Dankiewicz, G.F. Koob, Opiate withdrawal-induced

fos immunoreactivity in the rat extended amygdala parallels the

S. Liu et al. / Brain Research 1044 (2005) 155–163 163

development of conditioned place aversion, Neuropsychopharmacol-

ogy 24 (2001) 152–160.

[19] J.S. Han, Acupuncture: neuropeptide release produced by electrical

stimulation of different frequencies, Trends Neurosci. 6 (2003)

17–22.

[20] J.S. Han, R.L. Zhang, Suppression of morphine abstinence syndrome

by body electroacupuncture of different frequencies in rats, Drug

Alcohol Depend. 31 (1993) 169–175.

[21] S. Heinrichs, F. Menzhagi, G. Schulteis, G.F. Koob, L. Stinus,

Suppression of corticotrophin-releasing factor in the amygdala

attenuates aversive consequences of morphine withdrawal, Behav.

Pharmacol. 6 (1995) 74–80.

[22] N.H. Kalin, S.E. Shelton, R.J. Davidson, A.E. Kelley, The primate

amygdala mediates acute fear but not the behavioral and physio-

logical components of anxious temperament, J. Neurosci. 21 (2001)

2067–2074.

[23] G.F. Koob, Drug addiction: the yin and yang of hedonic homeostasis,

Neuron 16 (1996) 893–896.

[24] K.J. Kovacs, c-Fos as a transcription factor: a stressful (re) view from

a functional map, Neurochem. Int. 33 (1998) 287–297.

[25] J.E. Le Doux, Emotion circuits in the brain, Annu. Rev. Neurosci. 23

(2000) 155–184.

[26] P.H.K. Lee, R.W. Mcnutt, K.J. Chang, A nonpeptidic delta opioid

receptor agonist, BW373U86, attenuates the development and ex-

pression of morphine abstinence precipitated by naloxone in rat,

J. Pharmacol. Exp. Ther. 267 (1993) 883–887.

[27] S. Levine, H. Ursin, What is stress, in: M.R. Brown, G.F. Koob, C.

Rivier (Eds.), Stress, Neurobiology and Neuroendocrinology, Marcel

Dekker, New York, 1991, pp. 3–21.

[28] R. Maldonado, L. Stinus, L.H. Gold, G.F. Koob, Role of different

brain structures in the expression of the physical morphine withdrawal

syndrome, J. Pharmacol. Exp. Ther. 261 (1992) 669–677.

[29] A. Margolin, D.H. Kleber, S.K. Avants, J. Konefal, F. Gawwin, E.

Starkiff, E. Wells, T.R. Jackson, M. Bullock, P.D. Culliton, S. Boles,

R. Vaughan, Acupuncture for the treatment of cocaine addiction,

J. Am. Med. Assoc. 287 (2002) 55–63.

[30] K. Montazeri, M. Farahnakian, M. Saghaei, The effect of acupuncture

on the acute withdrawal symptoms from rapid opiate detoxification,

Acta Anaesthesiol. Sinics 40 (2002) 173–177.

[31] J.I. Morgan, T. Curran, Immediate-early genes: ten years on, Trends

Neurosci. 18 (1995) 66–67.

[32] G.A. Olson, R.D. Olson, A.J. Kastin, Endogenous opiates: 1989,

Peptides 11 (1990) 1277–1304.

[33] G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates,

Academic, London, 1986.

[34] P.V. Piazza, S. Maccari, J.M. Deminiere, M. Moal Le, P. Mormede,

H. Simon, Corticosterone levels determine individual vulnerability to

amphetamine self administration, Proc. Natl. Acad. Sci. U. S. A. 88

(1991) 2088–2092.

[35] T.L. Pierce, C. Raper, The effects of laboratory handling

procedures on naloxone precipitated withdrawal behavior in mor-

phine-dependent rats, J. Pharmacol. Toxicol. Methods 34 (1995)

149–155.

[36] L.J. Punch, D.W. Self, E.J. Nestler, J.R. Taylor, Opposite modulation

of opiate withdrawal behaviors on microinfusion of a protein kinase a

inhibitor versus activator in the locus coeruleus of periaqueductal

gray, J. Neurosci. 17 (1997) 8520–8527.

[37] C.N. Rosario, A.M. Pacchioni, L.M. Cancela, Influence of acute or

repeated restraint stress on morphine-induced locomotion: involve-

ment of dopamine, opioid and glutamate receptors, Behav. Brain Res.

134 (2002) 229–238.

[38] X.D. Shi, W. Ren, G.B. Wang, F. Luo, J.S. Han, C.L. Cui, Brain

opioid-receptors are involved in mediating peripheral electric stim-

ulation-induced inhibition of morphine conditioned place preference

in rats, Brain Res. 981 (2003) 23–29.

[39] M. Shwartz, R. Saitz, K. Mulvey, P. Brannigan, The value of

acupuncture detoxification programs in a substance abuse treatment

system, J. Subst. Abuse Treat. 17 (1999) 305–312.

[40] L. Stinus, M. Le Moal, G.F. Koob, Nucleus accumbens and amygdala

are possible substrates for the aversive stimulus effects of opiate

withdrawal, Neuroscience 37 (1990) 767–773.

[41] P. Veinante, M.E. Stoeckel, F. Lasbennes, M.J. Freund-Mercier, C-Fos

and peptide immunoreactivities in the central extended amygdala of

morphine-dependence rats after naloxone-precipitated withdrawal,

Eur. J. Neurosci. 18 (2003) 1295–1305.

[42] B. Wang, F. Luo, Y.Q. Xia, J.S. Han, Peripheral electric stimulation

inhibits morphine-induced place preference in rats, NeuroReport 11

(2000) 1017–1020.

[43] T. Watanabe, R. Yamamoto, A. Maeda, T. Nakagawa, M. Minami,

M. Satoh, Effects of excitotoxic lesions of the central or basolateral

nucleus of the amygdala on naloxone-precipitated withdrawal-induced

conditioned place aversion in morphine-dependent rats, Brain Res.

958 (2002) 423–428.

[44] H.K. Wen, W.K.K. Ho, Suppression of withdrawal syndrome by

dynorphin in heroin addicts, Eur. J. Pharmacol. 82 (1982) 183–186.

[45] H.L. Wen, J.C.K. Kwok, N. Datta, K.Y. Sin, S.K. Teoh, W.K.K. Ho,

Detoxification of heroin abusers by dynorphin (1–13), Pharmacol.

Commun. 1 (1992) 259–265.

[46] J.L. Williams, R.C. Drugan, S.F. Maier, Exposure to uncontrollable

stress alters withdrawal from morphine, Behav. Neurosci. 98 (1984)

836–846.

[47] L.Z. Wu, C.L. Cui, J.B. Tian, D. Ji, J.S. Han, Suppression of morphine

withdrawal by electroacupuncture in rats: dynorphin and kopioid

receptor implicated, Brain Res. 851 (1999) 290–296.

[48] G.D. Yang, W.H. Zhou, F.Q. Zhang, Experimental study of

selective muscarinic receptor antagonists on attenuation of morphine

tolerance and dependence in rats, Nat. Med. J. China 77 (1997)

130–133.